Bayesian uncertainty decomposition for hydrological projections
- PDF / 1,252,300 Bytes
- 23 Pages / 439.37 x 666.142 pts Page_size
- 35 Downloads / 175 Views
Online ISSN 2005-2863 Print ISSN 1226-3192
RESEARCH ARTICLE
Bayesian uncertainty decomposition for hydrological projections Ilsang Ohn1 · Seonghyeon Kim1 · Seung Beom Seo2 · Young-Oh Kim3 · Yongdai Kim1 Received: 6 November 2019 / Accepted: 16 December 2019 © Korean Statistical Society 2020
Abstract There is a considerable uncertainty in a hydrological projection, which arisen from the multiple stages composing the hydrological projection. Uncertainty decomposition analysis evaluates contribution of each stage to the total uncertainty in the hydrological projection. Some uncertainty decomposition methods have been proposed, but they still have some limitations: (1) they do not consider nonstationarity in data and (2) they only use summary statistics of the projected data instead of the full time-series and lack a principled way to choose the summary statistic. We propose a novel Bayesian uncertainty decomposition method which can alleviate such problems. In addition, the proposed method provides probabilistic statements about the uncertainties. We apply the proposed method to the streamflow projection data for Yongdam Dam basin located at Geum River in South Korea. Keywords Bayesian statistics · Heavy-tailed distribution · Hydrological projection · Uncertainty decomposition
1 Introduction It is well known that there is a considerable uncertainty in hydrological projections. A water management plan established without considering uncertainty can cause large losses if extreme events occur that are not anticipated by a hydrological projection. Therefore, it is recognized by many researchers and stakeholders that appropriately reflecting the uncertainty of water resources projection is an important task. Many
B
Yongdai Kim [email protected]
1
Department of Statistics, Seoul National University, 1, Gwanak-ro, Gwanak-gu, Seoul 08826, Korea
2
Water and Land Research Group, Korea Environment Institute, Seoul, South Korea
3
Department of Civil and Environmental Engineering, Seoul National University, Seoul, South Korea
123
Journal of the Korean Statistical Society
studies have reported uncertainties in hydrological projections for various regions (Lettenmaier and Gan 1990; Beven and Binley 1992; Kuczera and Mroczkowski 1998; Nijssen et al. 2001; Minville et al. 2008; Bastola et al. 2011; Nóbrega et al. 2011; Dobler et al. 2012; Mandal and Simonovic 2017). A hydrological projection consists of multiple stages including emission scenarios, global circulation models (GCMs), bias-correction techniques and hydrological models. Emission scenarios estimate future greenhouse gas emissions based on the set of assumptions about future energy use, technological change, population levels and economic activity. GCMs are numerical models that simulate various meteorological variables such as temperature and precipitation under the selected emission scenario. Bias correction techniques remove the systematic biases of the GCM outputs with respect to the observed data. Hydrological models produce projected values of a hydr
Data Loading...
